From his seat, David watched his breath fog in the crisp air of the converted barn. The rustic charm of the venue was lost on him. He felt the familiar constriction in his chest as he walked to the stage. His parents, seated in front, were staring expectantly, their faces filled with love.

He picked at the hem of his jacket. It was too tight, he thought. He felt pinned to the spot. The clinking of glasses and the murmur of conversation seemed to grow louder, amplifying every shaky breath he took. He'd rehearsed the speech a dozen times, but now the words had vanished, replaced by a suffocating silence.

He imagined turning and running, just disappearing out the back door, leaving the whole thing behind. But he couldn't.

Emotion: tense

Cluster: Fear / Anxiety
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -0.53

Role in Research

This story is one of 1,000 stories generated for the emotion tense. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 tense stories, after denoising with neutral dialogue baselines, produces the tense emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the tense vector is projected through the unembedding matrix.

Promoted:
S0.213
😰0.191
느껴0.183
😣0.183
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198